The Status of the “Material” in Theories of Culture: From ... · The Status of the “Material” in Theories of Culture: From “Social Structure” to “Artefacts” ANDREAS

The Status of the “Material” in Theories of Culture 195

© The Executive Management Committee/Blackwell Publishers Ltd. 2002© The Executive Management Committee/Blackwell Publishers Ltd. 2002. Published by Blackwell Publishers, 108 CowleyRoad, Oxford, OX4 1JF UK and 350 Main Street, Malden, MA 02148, USA.

Journal for the Theory of Social Behaviour 32:20021–8308

The Status of the “Material” in Theories ofCulture: From “Social Structure” to “Artefacts”

ANDREAS RECKWITZ

In the past several decades, social theory has been transformed largely intocultural theory. In different theoretical branches, the social has been redefined asthe cultural: Structuralism, semiotics and poststructuralism, phenomenology andhermeneutics, Wittgenstein’s language-game philosophy and symbolic interaction-ism have, in diverse ways, furthered a perspective that understands the orderli-ness of the social world as a result of symbolic structures. In contrast to classicaltypes of social theories such as naturalism, utilitarianism or the “homo sociologicus”of the norm-following actor, theories of culture can be defined as vocabulariesthat understand or explain human action and social order by establishing theirbasis in symbolic codes and schemes that regulate meaning. The classical dualismsof modern thought between “idealism” and “materialism”, between the realm ofthe “ideal” and that of the “real”, between the culture of the symbolic and thefactualism of material objects (dualisms which surely are not identical) thus ap-pear to have been resolved in favour of the former elements in these classicaloppositions. Now it seems as if—to borrow Derrida’s terminology (1967)—withinthe cultural/material distinction the material functions as the “supplément”, asthat element “added” to something already complete in itself: to culture.

However, this is the case only at first glance. Cultural theories have alwaystaken considerable trouble to answer the question of where to place the materialin relation to the symbolic. What is the status of the “material” dimension andhow is it defined within the vocabularies of the theories of culture? It must beemphasized that the idea of “materiality” does not have a common meaningamong theorists of culture, but that within these vocabularies it rather occupiesthe place of the “non-cultural”, which is conceptualized in very diverse ways. Wecan learn a great deal about theories of culture by seeking to reconstruct theplace where they localize these “non-cultural” elements which can be generallydescribed as “material”. And we can learn a great deal about the transformationthe theories of culture have experienced, and can experience, by following theshifted place and significance the “material” adopts within them. I propose to

JTSB32.2C04 5/27/02, 11:46 AM195

196 Andreas Reckwitz

© The Executive Management Committee/Blackwell Publishers Ltd. 2002

distinguish among three phases in the development of theories of culture thatdiffer in their conceptualization of the material: 1) the sociology of knowledgeas formulated by classical sociology in the work of Mannheim, Scheler andDurkheim; 2) “high modern” cultural theory as we find it in its different versionsin structuralism and social phenomenology (two variations of “culturalist ment-alism”), in poststructuralist and constructivist “textualism” and in Habermas’s“intersubjectivism”; 3) contemporary practice theory formulated in a radicalversion concerning the status of “artefacts” in the work of Bruno Latour. Only inthe third phase does it seem that cultural theory has reached a position capableof clarifying the relationship between the cultural and the material in a way thatis neither “culturalist” nor “materalist”. This ability is closely connected with thedevelopment of “theories of social practices” within the culturalist camp.

My argument will pursue the following line: 1) Classical sociology of know-ledge understands the material as “social structures” that provide a foundation fororders of knowledge. This approach emerges as a culturalist-materialist “dou-ble”. 2) We can understand the different branches of high-modern culturalism asa reaction against the insufficiencies of this vocabulary which stills seeks socialfoundations outside culture. High-modern culturalism redefines the material as“objects of knowledge” or “symbolic objects”, as objects which become visible in thecontext of systems of meaning (categories, discourse, communicative action). Inthe field of cultural theories of the last decades, this conceptualization of thematerial has been the dominant one. 3) An instructive novel conceptualizationof materiality within cultural theories is above all offered in the work of BrunoLatour (and in addition to certain other culturalist theorists of technology such asD. Haraway and F. Kittler). Bruno Latour’s “symmetric anthropology” can beunderstood as a critique of a reduction of social order to dematerialized symbolicorders and of the material to objects of interpretation. It enables one to grasp thematerial not as a social structure or as symbolic objects, but as “artefacts”, as“things” which are necessary components of social networks or “practices”.Latour’s sketchy approach, however, does not form an isolated position, but canbe embedded within the broader and more systematic movement of revisingcultural theory in the sense of a “theory of social practices”, as developed indetail by Theodore Schatzki.

1. CLASSICAL SOCIOLOGY OF KNOWLEDGE:

THE MATERIAL AS SOCIAL STRUCTURES

The classical sociology of knowledge, as we can find it in the work of KarlMannheim and Max Scheler in the 1920s and 1930s and—in a different form—in Émile Durkheim’s late work, occupies an ambivalent place in the landscapeof modern social theory. The sociology of knowledge provides an early version ofa theory of culture, and avant la lettre contributes to the “cultural turn” in social

JTSB32.2C04 5/27/02, 11:46 AM196



theory. Simultaneously, it relativizes this culturalist position by combining itsanalysis of collective stocks of knowledge with a materialist argument whichassumes that these symbolic structures “depend upon” social structures. In thesociology of knowledge, the place of the material is thus situated in social struc-tures outside culture; the material turns out to be the secret cause and foundationof the cultural. Classical sociology of knowledge thus represents a sort of culturalist-materialist “double”.1 However, the precise status of this alleged causal “link”between social structure and knowledge turns out to be doubtful.

For most “classical” social theorists who preceded the sociologists of know-ledge, the “social” was in no way identical with the “cultural”. This holds (withthe notable exception of Max Weber) for Karl Marx’s historical materialism, asit does for the early Durkheim’s theory of social evolution and social facts andGeorg Simmel’s “formal sociology” approach. Despite all the differences withinthese classical vocabularies the realm of “ideas”, the plane of cultural definitionsand interpretations becomes epiphenomenal; its analysis is not necessary in ex-plaining action and social order. The causal conditions of human behaviour andaction are rather situated on a different level: in the realm of pre-cultural “struc-tures”. It is contested among classical social theorists just how these social struc-tures are to be conceptualized: as productive forces and consequence laws (Marx),as structures of social differentiation, social density and social facts (Durkheim),or as social “forms” in contrast to social “content” (Simmel). It remains largelyuncontested, though, that the place of the social is social structure, and that socialstructure is a non-ideational sphere of regularities and patterns that exists andhas causal effects independent of subjective or collective interpretations.2

Against this backdrop, it is clear how the sociology of knowledge combines“new” and “old” conceptual elements. It breaks with classical social theory byunderstanding symbolic orders not as an epiphenomenon but as a necessary condition

of the orderliness of action. In this sense, we can read it as an early version of atheory of culture. At the same time, however, it takes over from its theoreticalpredecessors the definition of the material as a social structure that is situatedbeyond culture and that functions as the ultimate foundation of human action.Ergo, the sociology of knowledge is based on the idea of a “double-structure”forming the condition of action: The orderliness of action in social collectivitiesdepends on a shared cultural structure of knowledge. This one, for its part,depends on an objective “social structure”. Not social structure, but only thecultural structure of shared knowledge renders the structured nature of humanaction understandable—this is the culturalist argument in the sociology of know-ledge. Yet “in the last instance” the cultural structure is determined by a mate-rial social structure—this is the materialist argument, which the sociologists ofknowledge also support.

We can find a classical formulation of the culturalist-materialist “double” ofthe sociology of knowledge in Karl Mannheim’s Ideologie und Utopie (1929). Compar-able though not identical versions appear in Max Scheler’s Die Wissensformen und

JTSB32.2C04 5/27/02, 11:46 AM197



die Gesellschaft (1925) and in Émile Durkheim’s Les formes élémentaires de la vie

religieuse (1912). Mannheim reverses the classical anti-culturalism of 19th-centurysocial theory by shifting the meaning and status of the concepts of “ideology”and of “knowledge”. In the classical theories of ideology as they dominated 19th-century social thought, ideology was understood as “distorted knowledge”, asdelimited from “true”, i.e. proper knowledge. Classical analyses of ideology thenare based upon the traditional philosophical distinction between knowledge and“mere” belief (whose objective truth cannot be demonstrated). The sociologyof knowledge carries out a shift in the meaning and status of the concepts ofideology and knowledge, enabling the latter to become leading ideas of culturaltheory: Mannheim pursues the project of a “totalization” of the concept of ideo-logy that now embraces the whole of forms of knowledge as they are sharedby different social groups. Hence, he uncouples the concept of ideology fromthe question of truth or error and rejects the distinction between knowledgeand belief. There is no universal social-scientific catalogue of criteria left tojudge whether some forms of knowledge of certain social groups are invalid or“distorted”—rather, every form of knowledge seems to exist in relation to aspecific social group and can raise its truth-claim only relative to this communityof knowledge. (see Mannheim 1929: 49–94) Knowledge in its cultural contextualityis now regarded as a collective and constitutive background of patterns of action.Mannheim here leans on Husserl’s phenomenological idea that human actiontakes place before the “background” of a meaningful horizon that enables theagent to ascribe particular meanings to particular objects—the way of acting,then, depends on how meanings have been ascribed. This background is neces-sarily collective, not purely individual. Mannheim labels this background “Denkstil”(thought style) or “Aspektstruktur des Bewußtseins” (aspect structure of con-sciousness), while Durkheim talks about “systems of thought” or “collective ideas”.

The transformation of the concepts of ideology and knowledge enables thespecifically culturalist outlook of the sociology of knowledge. The sociology ofknowledge is culturalist insofar as it explains patterns of action in social collectivitiesby having recourse to the shared “thought style”, to the “collective representations”of a social group. However, Mannheim’s sociology of knowledge—like thoseversions of Scheler and late Durkheim—proceeds from the assumption that for asociological explanation of human behaviour such a culturalist pattern of explanationremains insufficent. The sociological explanation must search for a foundation ofcultural structures of knowledge in social structures that are themselves not idea-tional but material. Implicit in this line of argument is the idea that the specifically“social” foundation of human action cannot be cultural: the social is not thecultural but rather the material situated on the plane of “social structures”. Butwhat are social structures? And what is their “material” content? Mannheim(1929: 229–244) charges a number of different concepts with the task of para-phrasing these social structures: they embrace “conditions of being” (Seinslage),“conditions of structure” (Strukturlage) or “factors of being” (Seinsfaktoren)3.

JTSB32.2C04 5/27/02, 11:46 AM198



Max Scheler (1925: 21ff ) prefers to use the term “real factors” (“Realfaktoren”),distinguished from “ideal factors” (“Idealfaktoren”), to designate the non-ideal“basis” of culture. Émile Durkheim in his late work (1912: 13, 17, 21, 26) appliesa terminology of “collective realities”, “social conditions”, the “social origin ofcategories” and the “nature of things” to describe the social basis of collectiverepresentations (particularly of religious representations).

Why do sociologists of knowledge search for a basis of culture outside culture,in the “material world”? It seems that their fear of “relativism” provides the reason.In the moment the sociology of knowledge treats collective knowledge not as anepiphenomenon but as a necessary condition of patterns of action, it must on afirst level concede that all schemes of knowledge are “equal”, even if the beliefsthey back contradict one another. When the beliefs are “true” from the point ofview of the respective agents, they have effects on their actions. Yet such aradically contextualist statement leaves the sociologists of knowledge—Mannheimand Durkheim alike—uncomfortable. From their perspective, a sociologicalexplanation must reinterpret the seemingly “arbitrary” and contingent schemes ofknowledge as necessary phenomena on a more basic level—as necessary productsof certain social structures. In order to prevent the cultural knowledge of acollectivity from appearing as an entirely contingent phenomenon—which couldin principle always be otherwise—the sociology of knowledge realizes an escape.The emergence of a certain cultural knowledge in a collectivity requires itself anexplanation—one that makes it possible to derive the form of knowledge fromthe form of a more basic plane of phenomena: the pre-cultural plane of socialstructures.4

What, however, is the exact logical status of the relations between collectiveknowledge and “social structure”? In the sociology of knowledge, the sphere ofthe action-directing stocks of knowledge and the sphere of the more basic “socialconditions” and “real factors” are conceptualized as two distinct realms or levels.Here we can find three different versions of how the logical relation betweenthese two realms is conceptualized: as a relation of “causality”; as a relation of“reflection”; or as a relation of “social constitution”. Correspondingly, the idea of anon-cultural social structure differs. In the three cases, different problems ofconceptualization arise.

The first possibility of how to understand the link between social structure andculture/knowledge can be found in parts of Mannheim and Scheler. Here, the“conditions of structure” are presented as a cause for the emergence and repro-duction of the respective “thought styles” that consequently appear as “effects”of their social basis. The relation between the structures of being and knowledgeis not only one of “correlation” but also one of “determination”.5 The method of“relating” culture and structure, then, turns out to be a method of “explaining”the existence of a form of knowledge from the structural conditions of its emer-gence—structural conditions of a social class, a generation, an intellectual school,etc. (Mannheim 1929: 265). It is difficult to find a precise definition of “structural

JTSB32.2C04 5/27/02, 11:46 AM199



conditions” in this sense in the sociology of knowledge; though it does seem thatMannheim identifies them in part with a certain distribution of “resources”, forinstance between social classes. Then, the members of such a class, generation,intellectual school etc., pursue interests which result from this structural conditionand which result in a certain world-view. This version of the sociology of knowledgeobviously remains closely related to a classical pattern of “basis” and “superstructure”.

In Durkheim’s version of a sociology of knowledge, we encounter a secondvariant of the link between social structure and culture. Here, the relation inquestion is one of epistemological reflection. The categories a social group developsto classify social or natural phenomena—for instance concepts of time, space,causality etc.—are not arbitrary inventions; they do not “construct” a symbolicworld as a fiction that has nothing to do with the “real” social and natural world.On the contrary, in the epistemological sense they “correspond to” the structuresof the real, natural and social, world. In this sense, the collective classifications ofknowledge reflect not only social structures but also natural structures; theyreflect the “nature of things”. (Durkheim 1912: 23–28) If in the first version therelation between structure and culture resembles the classical link between basisand superstructure, then the second version implies similarities with the epis-temological relation between subject and object. The role of the classical subjectis now replaced by the symbolic categories of a social collectivity, while the roleof the classical object is occupied not only by natural, but also by social struc-tures. At any rate, the sociology of knowledge of the second version presupposesa primacy of the “object”, of the structures of the real world that “manifestthemselves in” social categories.

Finally, there is a third version in which sociologists of knowledge seek tograsp the relation between social structure and knowledge. It can be labelled themodel of social constitution, which we come across again in the work ofMannheim. In part, Mannheim’s argument of the “Seinsverbundenheit” of know-ledge does not mean that the realm of the cultural is “conditioned” by a realm ofthe social as the non-cultural, but rather that the stocks of knowledge are social-ized and transmitted in social processes, i.e. in processes of social interactionwithin groups. An agent does not learn his thought style in an individual con-frontation with a certain world of objects and events, but by being socialized intoa certain thought style which is defined as “true” and “normal” within a group,a social class or generation; knowledge is thus a product of a “social constitution”,of social conventions.6 These social processes in which agents develop their“forms of thinking” embrace not only processes of “taking over” categories fromothers, but also processes of delimiting oneself from the thought styles of certainothers (e.g. in intellectual competition). At any rate, the third version of theconceptualization of the relationship between social structure and culture modi-fies the meaning of “social structure” to a considerable degree (and in secret).“Social structure” loses its status as a distinct non-cultural and material realm,appearing instead to be a set of intersubjective processes in which collective

JTSB32.2C04 5/27/02, 11:46 AM200



knowledge is acquired. Thus “social structures” are converted from non-culturalpatterns to rules of knowledge-acquisition.

All three versions of linking social structure and culture raise their specificconceptual problems. The proposal that the relationship should be regarded as acausal link, collective knowledge treated as an “effect” and the non-cultural socialstructures as a “cause” leads to the question of which understandable mechanism isto connect cause and effect. After all, the sociology of knowledge assumes notonly a descriptive “correlation” between the observability of certain forms of know-ledge and the existence of certain social classes or other groups and their non-cultural resources; more than that, it assumes an explanatory logic between socialstructure and culture. To assume an explanatory logic one requires a mechanismthat elucidates why A “engenders” B.7 Chronic difficulties, however, remainwhen it comes to finding a “meaningful”, understandable “mechanism” mediatingbetween a non-cultural and a cultural realm. When the social structure of agroup is identified with its resources, and thus its status in a given society (thisalready presupposes a precise definition of social structure), it seems hardly pos-sible to explain why a certain material structure would “cause” a certain wayof interpreting the world and exclude other world-views.

Mannheim himself concedes that the classical notion of “interest” does notpromise any satisfactory solution here (1925: 375–387): Certain resources, aspecific social living condition or status do not “engender” any “corresponding”interest. Rather, which “interests” a social group pursues depends on how it defines

its interest, a definition rooted in the cultural realm of world-views and know-ledge itself.8 To interpret the link between social structure and culture as a relationof epistemological “reflection”, as in the second variant, amounts to a compar-able issue of how to maintain the alleged primacy of the “real” towards the“ideal”. When the sociology of knowledge contends that the “nature of things”—including “social things/facts”—“manifests itself” in the way these things areinterpreted and “experienced” in the collective categories of a group, it does notdeal with the “underdetermination of (everyday) theories by facts”.9 How is itpossible for a group to approach the “real” structures of the natural and socialworld when there are always potentially several ways of making sense of experi-ences, ways that are “empirically equivalent” and that cannot be situated in arelation of “correspondence” to social and natural reality? Again, the sociologyof knowledge does not suggest a mechanism according to which the “nature ofthings” is supposed to “manifest itself ” in the interpretation of the nature of things.

The third variant of conceptualizing the relation between social structure andculture, the pattern of “social constitution” as it is suggested partly by Mannheim,is not entangled in the problem of finding an understandable mechanism be-tween the two realms. After all, here the sociology of knowledge dissolves the strictseparation between two ontological spheres and “dematerializes” the status ofsocial structures. That cultural knowledge is socially constituted now means noth-ing more than the presupposition that the sharedness of knowledge is a result of

JTSB32.2C04 5/27/02, 11:46 AM201



social conventions and interactions that enable a collective socialization of mean-ing patterns. In this way, the sociology of knowledge dissolves its explanatoryclaim, its claim to explain culture by having recourse to the more stable groundof social structure. Instead, it now only spells out what has already been impliedin its basic culturalist position: Thought styles are not individual, but in differentsocial groups collectively shared; to have been shared, they obviously must havebeen “socialized” within the group. Consequently, as soon as the sociology ofknowledge formulates a tenable relation “between” culture and social structure,the distinction between the two realms collapses and it must give up its explanat-ory claim of a “Seinsverbundenheit” of culture. Instead of a materialist-culturalistdouble, in this third version the sociology of knowledge changes into a version ofpure culturalism: there is no distinct place left for the materiality of structuresoutside culture. The social cannot be found in the alleged solidity of a pre-cultural sphere, but in the collectivity of symbolic orders themselves. Thus, thesociology of knowledge itself provides the transition to a second outlook on thematerial which cultural theories can offer: here the material is presented not aspre-cultural social structures, but as carriers of cultural symbols.

2. MENTALIST, TEXTUALIST AND INTERSUBJECTIVIST CULTURALISM:

THE MATERIAL AS A WORLD OF OBJECTS OF KNOWLEDGE

The classical sociology of knowledge is conceptually instructive because it findsitself situated at an intersection of the materialist theories that preceded it andthe culturalist approaches which followed. The “cultural turn” in social theorythat has been taking place since the late 1960s, however, is hardly based uponthe sociology of knowledge. Rather, contemporary theories of culture are prod-ucts of schools of thought in 20th-century social philosophy that demarcate asharp break with all connotations of traditional schemes of “basis and super-structure”. Above all, structuralism, semiotics and poststructuralism, pheno-menology and hermeneutics, Wittgenstein’s philosophy of language-games andsymbolic interactionism provide different backgrounds for contemporary theor-ies of culture. Despite all the profound differences in their conceptualization ofthe social and of meaning, most of these cultural theorists share an overarchingposition vis-à-vis the status of the material world: The material world now nolonger adopts the status of a structural cause or condition of culture; instead,material entities appear as objects that gain a symbolic quality within classifica-tion systems, discourses or language-based interaction. The material world exists

only insofar as it becomes an object of interpretation within collective meaningstructures. There are no material entities as such, but only systems of distinctionsthat define certain “material objects” in a certain form and delimit them fromother (material or non-material) objects. Material entities exist as carriers ofmeaning, as “objects of knowledge”.

JTSB32.2C04 5/27/02, 11:46 AM202



It maybe astonishing to realize that in spite of the significant differences andcontroversies characterizing the field of contemporary cultural theories, almostall of them are based upon a tacit agreement as to the status of the materialas symbolic objects. Contemporary cultural theories ascribe no independentexplanatory force to a sphere of material entities. Rather, they unequivocallyformulate a logic of explanation that regards the structurality of human actionsas resulting from collective symbolic orders—these collective symbolic ordersnow are themselves the last “foundation” of the social world and cannot be derivedfrom any more basic “material” or “social” plane. For the new cultural theories,influenced by structuralism, semiotics, phenomenology, hermeneutics, language-game philosophy or symbolic interactionism, the social is not separate from thecultural but largely is the cultural, i.e. it is identical with symbolic orders. Wefind different forms of this understanding of material entities as cultural symbolsin four versions of contemporary cultural theory that differ according to thestatus ascribed to “symbolic orders”: in objectivist and subjectivist mentalism, intextualism and in intersubjectivism.

In the first and second cases, i.e. the two opposed versions of mentalism, thesphere of symbolic orders is situated in mind, in mental qualities which areinterpreted either as unconscious codes or acts of consciousness. Structuralismand social phenomenology present themselves as paradigmatic representativesof these two branches. In the third case, in the framework of textualism, sym-bolic orders are understood as discourses or “texts” outside mind; here, certainversions of poststructuralism and hermeneutics can be situated. Finally,“intersubjectivism”—paradigmatically represented by the work of Habermas—interprets social interactions and their linguistic basis as a “place” of symbolicorders.10 All approaches, however, contain traces of a post-Kantian dualism ofsubject and object—“only” that the subject is partly replaced by different “figures”,such as discourse or communicative action, which take over the status of thesubject. This does not, however, concern the status of the objects—understood asobjects of knowledge, of interpretation or of semiosis.

One of the roots and present branches of contemporary cultural theory thatleaves behind the division between culture and social structure is structuralism asit was classically formulated by de Saussure and applied to the social sciences byLévi-Strauss in a paradigmatic form. For structuralism, the place of symbolicorders is within the cognitive unconscious of mind. The plane of the social isidentical with the plane of collective mental qualities. These “systems of classi-fication”, “schemes” or “codes” determine what can ever become an “object”within language and action. In classical structuralism (unlike poststructuralism)the systems of signs, understood as systems of pairs of signifier and signified, arepurely mental structures. It is thereby only consistent that Lévi-Strauss deals with“material phenomena” by analyzing systems of classification, as in La pensée

sauvage (1962), or myths, as in Mythologies (1964ff ), by thus examining a symbolicand cognitive “logic” within systems of distinctions. “Totems”—Lévi-Strauss’s

JTSB32.2C04 5/27/02, 11:46 AM203



classical example from the former work—are reconstructed as a result of specificsystems of classifications in which distinctions from the realm of culture and(cultural) distinctions from the realm of nature are tied to one another. In acertain way one can interpret Lévi-Strauss’s structuralism as a theoretical heirof late-Durkheim’s strategy of analyzing “collective representions”, but largelywithout pursuing the project of explaining them by deriving them from a pre-cultural social structure.11

Within the culturalist camp, social phenomenology represents—in certainaspects—an antipode to structuralism. Social phenomenology, stemming fromHusserl and paradigmatically expressed in the work of Schütz, rejects the struc-turalist attempt to reconstruct meaning structures from an “objective perspect-ive” that passes over acts of consciousness. On the contrary, it situates meaningin the “subjective perspective”, in the way a consciousness ascribes meanings toobjects. Mental acts of consciousness stand in a relation of “intentionality” to theobjects of the world: they refer to them, and by referring to them render themmeaningful. These “objects” to which consciousness refers can be human agents,non-living objects or abstract entities. What matters are the meaningful systemsof typification employed by consciousness to arrange these objects in a certainway and make them appear “real” to the subject. In Schütz’s Strukturen der Lebenswelt

(1975: 62–87, 224–290), material objects, then, are largely interpreted as prod-ucts of systems of typification that arrange these objects according to the Kantianschemes of space and time. Although social phenomenology and structuralismconceptualize “mind” and “mental categories” in conflicting ways, they both sharea mentalist vocabulary that presupposes mental categories as the “inner” sourceof social order. The subjectivist mentalism of social phenomenology and theobjectivist mentalism of structuralism both imply a direct Kantian heritage whichin the case of structuralism is hinted at by Lévi-Strauss (1963) and which in thecase of phenomenology is explicit in Husserl (1956: 93–113). In this way bothstructuralism and phenomenology are based on Kant’s asymmetric distinctionbetween subject and object: irrespective of a real world of things an sich, the mentalacts and structures “constitute” their world of objects according to specific schemes.

While structuralism and social phenomenology directly follow the classicalKantian idea of a “knowing subject” and of “objects to to be known by thesubject” in two variations—though without the concept of a “transcendental”subject in the strict sense—the theoretical movements within contemporaryculturalism that can be labelled “textualism” and “intersubjectivism” stand inopposition to Kantian philosophy of the subject—but only upon first considera-tion. The movement towards “textualism” criticizes any attempt to localizesymbolic orders in mental structures, but situates them instead on the plane ofdiscourses, extra-mental symbols or texts. Michel Foucault’s “archaeology ofknowledge”, Clifford Geertz’s treatment of “culture as text” and Niklas Luhmann’stheory of social systems as environment-observing systems represent differentforms of culturalist textualism. Despite profound differences between Foucault’s

JTSB32.2C04 5/27/02, 11:46 AM204



poststructuralism, Geertz’s symbolistic hermeneutics and Luhmann’s theory ofcommunication, all these approaches share the position of locating symbolicorders not in mind, but on the level of extra-mental signs—be they linguistic ornon-linguistic.

This conceptual shift from “mental categories” to textual or discursive codes,however, does not lead to a fundamental revision of the status of material entit-ies. The role of the subject—structuralist or phenomenological—is replaced bydiscourse, symbols or communication (in the sense of Luhmann) but the materialobjects continue to appear as products of symbolic orders, as objects of know-ledge. Thus, in Foucault’s Les mots et les choses (1966) the objects of scientific studiesare interpreted as contingent results of historical knowledge codes. In Geertz’santhropological “thick descriptions”, material entities and events such as cock-fights, funerals or acts of winking gain their cultural relevance from their sym-bolic value, by “standing for” more or less abstract phenomena: social conflicts,religious values or a social relation (Geertz 1973). In Luhmann’s analyses of the“binary codes” and “semantics” employed in social systems, “objects” appearas constructions of system-specific “distinctions”: Non-communicative and non-mental entities form organic and psychic systems in the “environment” of thesocial (and the psychic), which are interpreted by social (and psychic) systemsin a specific way. (Luhmann 1984, 1990)12

If for the different branches of textualism social order is a result of sign-systems, then for intersubjectivism it is a result of symbolic interactions betweenagents. Culturalist intersubjectivism is paradigmatically formulated in JürgenHabermas’s theory of communicative action. From this point of view, symbolicorders take on the role of a structure of shared propositions and pragmatic rulesof language use, which provide a background for the interactions between subjects.Building on Popper’s distinction between three “worlds”, i.e. three ontologicalrealms, Habermas categorizes the plane of the material as “world one”, separatefrom the world two of the mental acts of subjects and the world three of sociallyshared semantic propositions: In their “objectively shared” semantic proposi-tions, agents make statements about the “world one” of material entities. Theseappear as an object of knowledge, as something “we talk about” in our symbolic-linguistic interactions. (Habermas 1981, I: 114–151)

Both textualism and intersubjectivism keep their distance from mentalist theor-ies of the subject and from a neo-Kantian localization of symbolic and thussocial order in mind. The role of mind is replaced by that of discourse/texts/symbols or by that of symbolic interactions respectively. However, the statusof the subject-object-distinction as a whole, understood in an abstract sense,is not concerned with this conceptual modification that refers to the “subject-pole”. Just as in the classical Kantian distinction, the material world is inter-preted as a matrix of “phenomena” that achieve their meaning through symbolicschemes, here through world-constructing codes or semantic propositions. Inthe four different branches of high-modern cultural theory, in structuralist and

JTSB32.2C04 5/27/02, 11:46 AM205



phenomenological mentalism, in textualism and in intersubjectivism, the materialworld thus appears as the plane of objects to be known or to be observed, to betalked about or to be interpreted, each time constructed by cultural codes.

The advantages and merits of these radically culturalist approaches are beyondall doubt: Unlike the classical sociology of knowledge and the materialistapproaches, they drop the idea of a sphere of the “pre-discursive” and its allegedfounding function, having already collapsed in the sociology of knowledge.Instead, the culturalists demonstrate how not only social and cultural entities butalso “material” entitites gain their meaning and significance only in certain systemsof differences or interpretative schemes which define them in contingent ways.However, the question arises which “costs” high-modern cultural theories bringabout. The main problem of these approaches seems to consist in a suspiciouslytraditional conceptualization of materiality as objects of knowledge. The ideathat material things exist for us only within contingent systems of difference,interpretative schemes etc. is here linked to the seemingly taken-for-grantedidea that these material things have primarily the status of objects which areobserved and interpreted. The issue is, however, whether such a classicalsubject-object-conceptualization gives just due to the significance which materialartefacts bear in the social world.

Above all three conceptual simplifications, which are linked to one another,can be the prize: human action is reduced to “intersubjective” interactions betweensubjects or to actions which make use of or are products of symbolic orders; social

orderliness is understood as an exclusive result of individual-transcending symbolicorders (languages, discourses etc.); social change in history is identical with a changeof cultural codes. The question is whether an alternative vocabulary can bedeveloped which remains within the field of cultural theory and thus retains itsinsights, while still being able to conceptualize materiality in a less intellectualistway than as objects of knowledge: to regard human activities “with things” not asan epiphenomenon in relation to acitivities with subjects or abstract entitites; toregard social orderliness also as a product of socially stabilized artefacts; and, finally,to regard social change in connection which changes of artefacts (without fallingback into a “materialist” theory). Already in the field of the “high-modernist”theories of culture there have been tendencies to go beyond a model of materialentitites as symbolic objects and to allude to the “constitutive” role of artefacts inthe social world. In Strukturen der Lebenswelt, Schütz and Luckmann in part beginto give their social phenomenology a “pragmatist” turn by pointing out the “recipeknowledge” needed to handle artefacts in everyday life (1975: 139ff ). Moving ina different direction, in his “genealogical” works Foucault begins to study thosesubject-forming institutional practices which are not exclusively “discursive”, butwhich make use of novel artefacts (architecture in prisons, biotechnologicaldevelopments in “bio-power” etc.; Foucault 1975, 1976). However, we can findan unequivocal attempt to formulate an alternative conceptualization of materialobjects within cultural theory in the work of Bruno Latour.

JTSB32.2C04 5/27/02, 11:46 AM206



3. SYMMETRIC ANTHROPOLOGY AND PRACTICE THEORY: THE MATERIAL

AS “ARTEFACTS” AND AS INTEGRAL COMPONENTS OF PRACTICES

Cultural theory does have an alternative to “culturalism” and its understandingof “materiality” as “objects of knowledge”, without having to fall back into thematerialist-idealist double of the sociology of knowledge. This is Bruno Latour’sbasic assumption. We can understand his works as pleading for such a “third”alternative (mainly in Latour 1991). Latour, however, is not the only author whotries to reconceptualize the social significance of objects as artefacts. There areFriedrich Kittler’s (1985) and Mark Poster’s (1995) approaches to a history ofsymbolic orders as history of technical media of communication, AndrewPickering’s (1995) science studies, Donna Haraway’s (1991) approach to “cyborgs”and their force to transform late-modern society, and in a certain way GillesDeleuze’s (1980) concepts of the organization of space by means of territorializa-tion and deterritorialization could already be mentioned here.

Bruno Latour’s argumentation is at first highly indebted to the sociology andanthropology of science as it spread in the 1970s in the wake of Kuhn’s cultur-alist perspective on “normal science” and of ethnomethodology. These “sciencestudies”—above all in the context of the Edinburgh School—appear at first asparadigmatic examples of the cultural turn and its basic assumption that mater-ial “facts” cannot exercise any independent causal influence on the social world,but that on the contrary, these “facts” are products of certain cultural codesand social practices (for instance in the scientific laboratory). Indeed, the anti-naturalist merits of this culturalist outlook on scientific practice remain valuablealso for Latour. Yet this cannot be the last word. The seeming radicalism of aculturalist perspective lessens when one recognizes that contemporary culturalismturns out to be one more version of the problematic “modernist constitution”,i.e. a modernist way of thinking that tends toward a “purification” of culturefrom nature (including both the organic and the physical world) and of naturefrom culture. This semantic and practical purification contradicts and at thesame time enables the fact that in particular modern times are witnessing anunprecedented expansion of hybrids, “quasi-objects” (Serres), non-human creat-ures that are neither pure nature nor cultural projections, but indispensable (bynow, innumerable) components of social “networks” or “practices”.

The significance of these quasi-objects—from the ozone hole to HIV, fromcomputers to genes—has been systematically misunderstood by the dominatingconceptual dualism of nature and culture/society and by the at first glancecontradictory, but in fact comparable strategies of “naturalization”, “socialization”and “discursivation” in philosophy and the social sciences. However, Latour’sapproach to what he calls a “symmetric anthropology” provides more than anattempt to offer a novel and more adequate perspective on the innumerablephenomena of modern science and technology and their social effects. Moregenerally, it can be understood as an effort to work out a new (“non-modernist”)

JTSB32.2C04 5/27/02, 11:46 AM207



vocabulary to grasp the “link” between the cultural and the material beyond theidea of a “constitution” in one way or the other. To achieve this aim, Latourinsists, the material world must be seen neither as a basic structure at the foun-dation of any culture and knowledge, nor as a matrix of symbolic objects on thescreen of the respective culture. Rather, it should be understood as “artefacts” or“things” that necessarily participate in social practices just as human beings do.To be sure, these things are “interpreted” by the human agents in certain ways,but at the same time they are applied, used, and must be handled within in theirmateriality. As things, they are not arbitrarily interchangable.

Latour takes some pains to demonstrate the different intellectual strategiesmodern thought has conceived to create two different “chambers”, one of asocial and cultural world, i.e. a realm of human subjects, and one of a materialworld, i.e. a realm of non-human objects, as it is recognized in natural laws bymodern natural science. Above all, three conceptual strategies have hithertodominated the scene of philosophy and the social sciences (Latour 1991: chap.2). The first pattern is an explicit dualism between nature and culture, betweensubjects and objects as two realms existing independently of one another or evenin a relationship of “incommensurability”. This point of view sees “things” aselements of nature existing entirely separate from social relations. The secondpattern—according to Latour, particularly widespread in the social sciences—amounts to a relation of mutual constitution of culture/society and nature: Someelements of nature—those seen as “hard” and insurmountable “structures”—areinterpreted as a causal foundation of culture; simultaneously, certain otheraspects of the material world have effects “only” in the form of cultural interpre-tations. The third pattern carries out a resolute “discursivation” of the materialworld. This strategy can amount either to a radical “cultural relativism” thatassumes that every culture disposes of its own “picture” of nature, or to a par-ticularist universalism that assumes that though every culture fabricates its viewof nature, some views (normally those fabricated by Western science) are “moreadequate” than others. At any rate, the material entities do not appear as thingsto be handled, but as objects to be interpreted.

In relation to these conceptual purifications, Latour presents an alternativevocabulary which, however, appears in his works only in outline. The centralconcept of this alternative outlook is that of a “network” or a “practice” and ofa “nature/culture” (Latour 1991: chaps 3, 4). Neither the material world nor asocial world of meanings or power relations, can be taken for granted as aseemingly certain point of departure. Instead of following anthropology orculturalist sociology in distinguishing different “cultures” and their forms ofinterpretation, one should distinguish between different “natures/cultures”: Socialnetworks or practices in their historical variability consist not only of humanbeings and their “intersubjective” relationships, but also simultaneously of non-human “actants”, things that are necessary and are so-to-speak “equal” compon-ents of a social practice. In the presence of the explosion of technical artefacts

JTSB32.2C04 5/27/02, 11:46 AM208



in contemporary societies, it becomes more and more difficult to follow themodernist constitution and to overlook this constitutive status of things for socialpractices.

Yet this is already valid for so-called “pre-modern” societies (a notion Latourseeks to avoid), which only possessed considerably smaller numbers of sociallyrelevant things. “Things” thus have the status of “hybrids”: On the one hand, theyare definitively not a physical world as such, within practices they are sociallyand culturally interpreted and handled. On the other hand, these quasi-objectsare definitively more than the content of cultural “representations”: they are usedand have effects in their materiality. Without the specific materiality of a vacuumpump the respective scientific practice would not be possible—the vacuum pumpcannot simply be replaced by some other arbitrary “symbolic object” to which thesame “meaning” is then ascribed. Hence, the social scientist cannot confine him-self to the analysis of “cultures” (and in fact, at least the anthropologist who wasconcerned with pre-modern societies has never taken this limitation seriously).Rather, networks of human beings, of non-human creatures and their regulatedrelationships to one another, form historically specific “natures/cultures”.

For Latour this rehabilitation of the material world in the form of artefacts,integrated in social practices, amounts to a novel concept of the “collectivity” andto a dissolution of the micro/macro-distinction notorious in modern socialtheory. Classically, social theory has confidently defined the “collective” as rela-tionships between human beings, between subjects, thus to found it in a constel-lation of “intersubjectivity”. The paradigmatic situation of “sociality”, then, usedto be that of a face-to-face-interaction between two or more agents. The tradi-tional critique reproached this sociological interactionism for ignoring those“social structures” that exist beyond the particular context in which the interac-tion takes place. But what are these social entities crossing different contexts andleading to a transcontextual social order? For Latour, referring to “symbols” ormeaning structures alone cannot offer a satisfactory answer to this question.Indeed, when comparing human societies to simian societies, the particularityof the former consists in the fact that they are more than just an agglomerationof social interactions under the condition of copresence. Rather, the stability of“human” social orders beyond particular contexts of action can be properlyunderstood only when one regards practices not merely as constellations ofintersubjectivity, but also as constellations of “interobjectivity” (Latour 1996):“By dislocating interaction so as to associate ourselves with non-humans, we canendure beyond the present, in a matter other than our body, and we can interactat a distance. . . . The old difference of levels comes merely from overlooking thematerial connections that permit one place to be linked to others and from beliefin purely face-to-face interactions.” (1996: 239f ) For Latour, the distinction be-tween micro-interactions and macro-structures thus dissolves itself: What we canfind is nothing more nor less than the “flat” level of social practices. Yet, in thesepractices, material things are routinely drawn upon and applied by different

JTSB32.2C04 5/27/02, 11:46 AM209



agents in different situations. The objects handled again and again endure, thusmaking social reproduction beyond temporal and spatial limits possible.

Bruno Latour does not, however, present his new outlook on the status ofmaterial objects in the form of an elaborated social theory and his approachcontains a number of conceptual ambiguities, among which the alleged statusof objects as “actants” in their own right is one of the most problematic andcontested (cf. the critiques in Collins/Yearley 1992, Gingras 1995, Bloor 1999).Instead of formulating a “Latourian” social theory—a project which should beseen with some scepticism, as his vocabulary is fixed on the nature/cultureproblem and merely touches upon many other significant social-theoretical issues(body/mind, understanding, social institutions etc.)—it seems more fruitful toinstrumentalize Latour’s ideas and to situate him in the broad stream of recentcultural theories which can be described as “theories of social practices”. Herewe encounter an interesting constellation: Systematic conceptualizations of apractice theory—as we can find it in different versions in certain works of AnthonyGiddens (1984), Pierre Bourdieu (1972) and Laurent Thévenot (2002), in sketchierversions in ethnomethodology (Garfinkel 1967) or the work of Judith Butler(1990), and in a social-philosophical form in Theodor Schatzki’s Social Practices

(1996)—have carried out the shift from “subjects” and “texts” to “social prac-tices” in a more complex way than Latour, but they have not or only to a limiteddegree rethought the status of material objects. Yet the idea of “practice theory”is not only compatible with, but demands such a rethinking. Conversely, Latour’ssketchy outline of a reconceptualization of the dualism between the cultural andthe material becomes more comprehensive—and is reread in a more “anthro-pocentric” way, as Latour probably would prefer it—once it is embedded in thebroader theoretical frame of a “theory of social practices”. Latour’s leadingconcept of social “networks” of “practices” consisting of both human and non-human creatures and their relationships can be reformulated when linked to“praxeological” thinking. The central issue then is that certain things or artefactsprovide more than just objects of knowledge, but necessary, irreplacable compon-ents of certain social practices, that their social significance does not only consistin their being “interpreted” in certain ways, but also in their being “handled” incertain ways and in being constitutive, effective elements of social practices.

Apart from textualism in poststructuralism, radical hermeneutics and theconstructivism of social systems, and apart from intersubjectivism in the form ofHabermas, practice theories provide an alternative attempt within contemporarycultural theory to overcome the model of the subject or mind as locus of thesocial and of knowledge. Whereas textualism and intersubjectivism both turn outto be secret allies of mentalism as far as their implicit subject-object dualism isconcerned, this is not the case for practice theory. But what is “practice theory”?13

It is a tautology to state that for practice theory the place of the social is neither inthe mind nor discourses nor symbolic interactions, but in “social practices”. Whatis a “social practice”? In Social Practices. A Wittgensteinian Approach to Human Activity

JTSB32.2C04 5/27/02, 11:46 AM210



and the Social (1996), Theodore Schatzki develops a detailed account of the philo-sophically sophisticated ideas of a theory of social practices, which in its system-atic character surpasses the accounts of the other relevant authors. This approachis based to considerable degree on elements of Wittgenstein’s analysis of language-games and of Heidegger’s analysis of “Dasein”. A central problem of social theoryhas hitherto consisted in its reification of the subject and of (collective) humanmind as its point of departure for social analysis. According to Schatzki, how-ever, the proper site of the social is not collective mind but social practices.

A social practice is a regular bodily activity held together by a socially stand-ardized way of understanding and knowing. A social practice of x-ing (cooking,excusing, researching, working, arguing etc.) is then “a temporally unfolding andspatially dispersed nexus of doings and sayings” (Schatzki 1996: 89), “organized”by a socially typical “understanding of x-ing” above all including practical know-ledge (apart from norms and teleoaffective structures). Schatzki criticizes allattempts—for instance in Lyotard’s work—to reduce social practices to discurs-ive practices. Practices are not only forms of “saying”—hitherto falsely privilegedin social theory—but also forms of doings. Doings and sayings cannot be con-ceptualized as mental competences or as sequences of signs, but present them-selves first and foremost as certain regular bodily activities. To form a sociallyconventionalized “intelligible” practice, however, these bodily activities need tobe “organized” by knowledge. This knowledge is necessarily expressed in bodilyactivities: “Mind . . . is the expressed of the body” (Schatzki 1996: 53). Mindcannot completely be excluded from social analysis, but it must be demystified:Regular bodily activities are necessarily accompanied by typical (mentally an-chored) forms of understanding and knowing (otherwise the body would not actregularly), but this understanding and knowing “exists” only insofar as it manifestsitself in bodily behaviour. When we talk of “social fields” or “institutions”, in theend we find nothing more than nexuses and sequences of social practices. The“symbolic orders’ highlighted by cultural theorists, then, turn out to be forms ofpractical understanding which organize practices. For practice theory, the small-est unit of social analysis can definitively not be found in qualities of the “subject”.But here the decentering of the subject is carried out in a form different fromtextualism or intersubjectivism. “Language” and “sign systems” as “supraindividual”or “intersubjective” entities are no longer the locus of the social; repetitive bodilyactivities, in which certain forms of understanding are expressed, take on this role.

Surprisingly, however, “things” are largely missing in Schatzki’s practice theory.In only one passage does he mention that “places are anchored in objects whichare combined into settings” (Schatzki 1996: 189). Thus, objects are presentedhere as symbolic markers which establish certain settings for certain practices.However, in accordance with the common culturalist marginalization of thepractical relevance of things, this position does not take full advantage of theconceptual chances offered by practice theory.14 If Schatzki emphasizes thatpractices are a nexus of “doings and sayings” and that they are not identical with

JTSB32.2C04 5/27/02, 11:46 AM211



constellations of intersubjectivity, then these doings must almost necessarily bedoings with things. It is possible that not all practices are doings with things, butmost practices (including Schatzki’s examples) are. Here we can integrate—atleast to a certain extent—Latour’s position in Schatzki’s: not only human beingsparticipate in practices, but also non-human artefacts form components of prac-tices. The things handled in a social practice must be treated as necessary com-ponents for a practice to be “practiced”. In fact, one can say that both the humanbodies/minds and the artefacts provide “requirements” or components neces-sary to a practice. Certain things act, so to speak, as “resources” which enableand constrain the specificity of a practice.15 Recent studies on communicativemedia, for instance—above all those by Friedrich Kittler—abound with examplesof how communicative practices change with the development of new mediain their technical materiality (writing, printing, audiovisual media, computer andInternet). These media are not mere instruments to “transmit” messages, butmould forms of perception and communication (see Kittler 1985, Gumbrecht1988, and Benjamin as early as 1936). Social change is thus more than exclusivelya change of cultural codes, but depends also on a change of technical media.

Yet technical equipment cannot determine certain activities in a strict causalway. In order to have effects, artefacts must be used; and to be used, they mustbe treated with understanding and within the parameters of cultural codes—theymust become an integral part of a social practice. Thus, from the point of viewof practice theory, the “relationship” between human agents and non-humanthings in the network of a practice is a relationship of practical understanding.Simultaneously, in such a relationship the artefacts do not allow any arbitrarypractical use and understanding, they are not suitable for arbitrary practices. IfSchatzki and practice theory emphasize that practices are organized by a certainway of understanding and knowing (primarily but not exclusively through prac-tical know-how), and if practices are not only routinized forms of saying, but alsoroutinized forms of doing, then the relationship between human agents andthings to be handled presupposes a practical understanding. When human agentshave developed certain forms of know-how concerning certain things, thesethings “materalize” or “incorporate” this knowledge within the practice (the latterrestriction is important because “as such” and beyond complexes of practicesthings do not incorporate anything—at least from the point of view of a post-Wittgensteinian theory). Things are “materialized understanding”, and only asmaterialized understanding can they act as resources. Practice theory in the mostsophisticated version offered by Schatzki stresses the bodily basis of all practiceson the part of human beings. In a very basic sense, all social practices arecollective and routinized movements of bodies. Thus, bodies are the site ofunderstanding, of “embodied understanding”. Yet in order to avoid the anti-praxeological schism between the cultural and the material, it must be stressedthat within practices not only bodies but also artefacts are sites of understanding,in the form of materialized understanding. Not only if the body/mind that

JTSB32.2C04 5/27/02, 11:46 AM212



“incorporates” and “embodies” a certain understanding disappears does it be-come impossible for a social practice to be reproduced. If the things that incor-porate a certain materialized understanding were to disappear (for instance certaincommunication media) or had never appeared, we would observe the same result:the impossibility of maintaining a certain social practice.

If we combine practice theory in the sense of Schatzki and Latour’s approachto artefacts, social order does not appear as a product of “symbolic orders” inminds, discourses or interaction. Practice theory follows the other types ofcultural theory in proceeding from the assumption that social order is formed inthe crucible of cognitive-symbolic relations. Yet, these cognitive-symbolic rela-tions can now be conceptualized as practical understanding that is—as Schatzkihighlights—incorporated in active bodies and—as Latour stresses—simultan-eously materialized in artefacts. The materiality of artefacts influences—but doesnot determine—which practical understanding and, consequently, which kindsof social practices are possible. In the form of things to be handled in the contextof practices, the material world is more than a matrix of symbols, but less thana “basis” for a cultural “superstructure”. Social order and reproduction can beadequately understood only when we realize their double localization: as under-standing incorporated in human bodies and as understanding materialized inartefacts. Thus in relation to “high-modern” cultural theories, practice theories,which generally reconceptualize human action, social orderliness and socialchange, can reconceptualize these phenomena also in relation to things. Social

orderliness is not localized in mental structures, discourses or intersubjectivity,but in the social practices for which human bodies/minds and artefacts formnecessary components. Consequently, within the field of human action, actionsbetween subjects lose their omnipotent status and actions between human agentsand non-human artefacts are rehabilitated in their significance (cf. also Knorr-Cetina 1997). Finally, if social change is a change of complexes of social practices,it presupposes not only a transformation of cultural codes and of the bodies/minds of human subjects, but also a transformation of artefacts (a relationshipwhich deserves closer study than can be offered here).

Of course, when choosing as our point of departure practice theory in thesense of Schatzki and in the historical tradition of Wittgenstein’s late andHeidegger’s early social philosophy16, while at the same time “integrating” Latour’sideas concerning the status of the material into this framework, we arrive at aversion of practice theory which requires further elaboration. One of the majorissues is to what extent such an “integration” and “instrumentalization” of Latourinto practice theory is possible and to what extent there remain considerabletheoretical differences between a practice theory which is modified along lines ofa theory of artefacts on the one hand and Latour’s “symmetric anthropology” onthe other hand. It seems that post-Wittgensteinian theory of social practices hasgood reason to regard artefacts as necessary and influential components of socialpractices, while wishing to retain an “asymmetric” relation between them and

JTSB32.2C04 5/27/02, 11:46 AM213



the human agents. When artefacts can only be effective within practices insofaras they are “handled” by human agents and when they are sites of “materializedunderstanding”, then their status obviously cannot be completely “equal” withthat of human agents and their embodied understanding. The distinction be-tween such a position and Latour’s pleading for a “symmetric anthropology”should not be blurred; rather, the debate whether within social practices thereis or is not any substantial difference between human agents and non-human“actants” must continue.17 In any case, such a debate would carry the con-ceptualization of the material considerably further than has been done by thesociology of knowledge and classical culturalism.

Andreas Reckwitz

Institut für Soziologie

Allende-Platz 1

20146 Hamburg

Germany

[email protected]

NOTES

1 I am adopting the metaphor of a “double” from Foucault who describes the para-doxical structure of Kant’s philosophy of the subject as a “doublet empirico-transcendantal”(1966: 329).

2 This identification of the social with social structures in the sense of regular, non-subjective patterns is today supported by authors such as Peter Blau and Jonathan Turner(cf. Turner 1987).

3 To understand the term “being” in this context, one should be aware of the con-notation of the German term “Sein”. Obviously Mannheim here builds upon Marx’sdistinction between “Sein” and “Bewusstsein”, i.e. being and consciousness. Thus here(unlike in Heidegger’s terminology) the term “being” does not have any “existential” con-notation. Consequently, it seems hardly justifiable to translate Mannheim’s “Seinsverbun-denheit” as “existential connectedness”, as it is done in the English translation of Ideologieund Utopie.

4 Mannheim describes his fear of relativism in detail. See Mannheim 1929: 38ff.5 For instance, see “zu zeigen, . . . daß diese . . . Seinsfaktoren . . . alles, was wir als

Aspektstruktur einer Erkenntnis bezeichnen werden, entscheidend bestimmen.” (Mannheim1929: 230)

6 For instance, see: “Auch wo es für das erlebende Subjekt den Anschein hat, alskämen seine “Einsichten”, “Intentionen” “einfallsmäßig”, sprunghaft und nur aus ihm, sostammen sie dennoch aus einer auch in ihm lebendigen . . . kollektiven Grundintention.”(Mannheim 1984: 68)

7 Using Max Weber, Mannheim himself points out that the sociological explanantionrequires “causal adequacy” (Kausaladäquanz) between A and B, but also an “adequacy ofmeaning” (Sinnadäquanz) between these two elements (Mannheim 1984: 56–59).

8 Very similar problems arise for Pierre Bourdieu, who in La distinction (1979) attemptsto present the emergence of the cultural schemes of the habitus as en effect of a certainstructure of “capital”.

JTSB32.2C04 5/27/02, 11:46 AM214



9 I borrow this phrase from W.V.O. Quine and his critique of empiricism (see Quine1960: chapter II).

10 The fifth version of contemporary cultural theory, practice theory, provides thesubject of part three of this article. For a more detailed comparison of the differentbranches of a theory of culture and a systematic reconstruction of their development, seeReckwitz 2000, 2002.

11 However, the work of Lévi-Strauss contains some relics of the basis/superstructure-division (1962: 247, 263f ). Here Lévi-Strauss tends to identify the cultural unconscious ofmind with the neuronal structures of the brain.

12 Luhmann’s concept of environment-observing social systems is built in formalaccordance with Husserl’s concept of an intentional consciousness. Now social systems,i.e. sequences of communication, are said to observe (i.e. construct) their environment inthe same way as Husserl’s consciousnesses do.

13 As to “practice theory” in general and the transformation of contemporary theoriesof culture towards practice theory both in the context of structuralism and interpretativetheories, see Reckwitz 2000, 2002.

14 Later on, Schatzki has developed an interest in artefacts, cf. Schatzki 2001. More-over, Schatzki’s new book, The Site of the Social to be published by Penn State Press in2002, promises to focus on these questions.

15 Here it is possible to create a link with Anthony Giddens’s version of practice theory,presented in the form of his “theory of structuration”. In Giddens’s conceptual framework“the material” appears as “resources”, which are interpreted as necessary requirementsfor the existence of practices. These resources, however, are primarily understood asallocative or authoritative means of power, less as things/artefacts to be handled. (seeGiddens 1984: 58–62) Nevertheless, it could be fruitful to create a connection betweenthe concept of artefacts and that of resources.

16 It would be particularly interesting to read Heidegger’s Sein und Zeit (1927) as anearly attempt to formulate a practice theory that rehabilitates the status of “things” withinthese practices. It would then come as no surprise that Heidegger does not start hisanalysis of human practice by studying “intersubjectivity”, but by having a closer look atthe “interobjectivity” of human “Dasein” and of artefacts in the mode of “Zuhandenheit”(see Heidegger 1927: 63–76). Within the specific field of the philosophy of technics, DonIhde in particular has followed a similar path (see Ihde 1979).

17 Many of the contributions in Schatzki/Knorr-Cetina/Savigny 2001 can be seen asstarting points for such a debate.

REFERENCES

B, W. (1936). Das Kunstwerk im Zeitalter seiner technischen Reproduzierbarkeit. Frankfurt/Main 1963: Suhrkamp.

B, D. (1999). “Anti-Latour”, Studies in history and philosophy of science, No. 1, 81–112.B, P. (1972). Esquisse d’une théorie de la pratique, précédé de trois études d’éthnologie kabyle.

Geneva: Droz.B, P. (1979). La Distinction. Critique sociale du jugement de goût. Paris: Minuit.B, J. (1990). Gender Trouble. London: Routledge.C, H.M. and S. Y (1992). Epistemological Chicken pp. 301–326 in A.

Pickering (ed.) Science as Practice and Culture. Chicago 1992: Chicago Univ. Press.D, G/F G (1980). A Thousand Plateaus. London 1988: Athlone.D, J. (1967). De la grammatologie. Paris: Minuit.D, É. (1912). Les formes élémentaires de la vie religieuse. Paris: Alcan.

JTSB32.2C04 5/27/02, 11:46 AM215



F, M. (1966). Les mots et les choses. Paris: Gallimard.F, M. (1975). Surveiller et punir. La naissance de la prison. Paris: Gallimard.F, M. (1976). Histoire de la sexualité I: La volonté de savoir. Paris: Gallimard.G, H. (1967). Studies in Ethnomethodology. Cambridge 1984: Polity.G, C. (1973). The Interpretation of Cultures. Selected essays. London 1993: Fontana.G, A. (1984). The Constitution of Society. Outline of the theory of structuration. Cambridge:

Polity.G, Y. (1995). “Un air de radicalisme. Sur quelques tendances rècentes en sociologie

de la science et de la technologie”, Actes de la recherche en sciences sociales, No. 6, 3–17.G, H.-U. (ed.) (1988). Materialität der Kommunikation. Frankfurt/Main: Suhrkamp.H, J. (1981). Theorie des kommunikativen Handelns. Frankfurt/Main, vols. 1+2:

Suhrkamp.H, D.J. (1991). Simians, Cyborgs, and Women. The reinvention of nature. New York/

London: Routledge.H, M. (1927). Sein und Zeit. Tübingen 1986: Niemeyer.H, E. (1956). Die Krisis der europäischen Wissenschaften und die transzendentale Phänomenologie.

Hamburg 1992: Meiner [Gesammelte Schriften 8; written in 1936/38].I, D. (1979). Technics and Praxis. Dordrecht: Reidel.K, I. (1781). Kritik der reinen Vernunft. Frankfurt/Main 1977: Suhrkamp.K, F. (1985). Aufschreibesysteme 1800/1900. Munich: Fink.K-C, K. (1997). “Sociality with objects. Social relations in postsocial knowl-

edge societies”, Theory, Culture & Society, No. 4, 1–30.L, B. (1991). Nous n’avons jamais été modernes. Essai d’anthropologie symétrique. Paris: La

Découverte.L, B. (1993). La clef de Berlin. Paris: La Découverte.L, B. (1996). “On Interobjectivity”, Mind, Culture, and Activity, No. 4, 228–244.L-S, C. (1962). The Savage Mind. London: Weidenfeld.L-S, C. (1963). “Réponses à quelques questions”, Esprit, 628–653.L-S, C. (1964). Mythologiques I. Le cru et le cuit. Paris: Plon.L, N. (1984). Soziale Systeme. Grundriß einer allgemeinen Theorie. Frankfurt/Main:

Suhrkamp.L, N. (1990). Die Wissenschaft der Gesellschaft. Frankfurt/Main: Suhrkamp.M, K. (1925). “Das Problem einer Soziologie des Wissens”, pp. 308–387 in K.H.

Wolff (ed.) Wissenssoziologie. Darmstadt/Neuwied 1970: Luchterhand.M, K. (1929). Ideologie und Utopie. Frankfurt/Main 1985: Klostermann.M, K. (1984). Konservatismus. Ein Beitrag zur Soziologie des Wissens. Frankfurt/Main:

Suhrkamp [written in 1925].P, A. (1995). The Mangle of Practice. Time, agency, and science. Chicago: Univ. of

Chicago Press.P, M (1995). The Second Media Age. Cambridge: Polity.Q, W.V.O. (1960). Word and Object. Cambridge (Mass.): MIT Press.R, A. (2000). Die Transformation der Kulturtheorien. Zur Entwicklung eines Theorieprogramms.

Weilerswist: Velbrück.R, A. (2002). “Toward a theory of social practices. A development in culturalist

theorizing”, European Journal of Social Theory, No. 2, 245–265.S, T.R. (1993). “Wittgenstein + Heidegger on the stream of life”, Inquiry, No. 3,

307–328.S, T.R. (1996). Social Practices. A Wittgensteinian approach to human activity and the social.

Cambridge: CUP.S, T.R. (2001): Introduction: practice theory pp. 1–14 in: Schatzki/Knorr-Cetina/

Savigny (2001).

JTSB32.2C04 5/27/02, 11:46 AM216



S, T.R., K K-C and E S (eds.) (2001). The PracticeTurn in Contemporary Theory. London: Routledge.

S, M. (1925). Die Wissensformen und die Gesellschaft. Bern/München 1960: Francke.S, A./T. L (1975). Strukturen der Lebenswelt, vol. 1. Frankfurt/Main:

Suhrkamp.S, G. (1908). Soziologie. Untersuchungen über die Formen der Vergesellschaftung. Frankfurt/

Main 1992: Suhrkamp.T, L. (2002). Sociologie pragmatique. Les régimes d’engagement. Paris (forthcoming).T, J (1987). “Analytical Theorizing”, pp. 156–194 in A. Giddens/J. Turner

(eds.) Social Theory Today. Cambridge 1987: Polity.W, M. (1922). Wirtschaft und Gesellschaft. Grundriß der verstehenden Soziologie. Tübingen

1980: Mohr.W, L. (1953). Philosophische Untersuchungen. Frankfurt/Main 1984: Suhrkamp

[Werkausgabe, vol. 1; written in 1945–49].

JTSB32.2C04 5/27/02, 11:46 AM217

Documents

The Status of the “Material” in Theories of Culture: From ... · The Status of the “Material” in Theories of Culture: From “Social Structure” to “Artefacts” ANDREAS